IBM_granite 4.1 30b_Abliterated needs about 20.1 GB of memory at Q4_K_M (a 17.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 24 quants, from 0.0 GB to 23.7 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.
IBM_granite 4.1 30b_Abliterated – a 30B Granite model with orthogonalized refusal removal
IBM_granite 4.1 30b_Abliterated is a 30B parameter Granite model with a context length of 131072 tokens. It removes safety guardrails via orthogonalization while preserving core knowledge. The model is released under the apache-2.0 license.
Good at
- Large context handling
- General language tasks
- High token throughput
Limits
- No vision support
- Low censorship level may produce unsafe content
- Requires large VRAM for higher quants
Best for
- Research on refusal mechanisms
- Testing safety removal techniques
- High‑capacity text generation
Who it’s for. Researchers and developers needing a large Granite model without safety guardrails. Suitable for experimentation with orthogonalization techniques and for applications that can handle low censorship levels.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.0 GB | 2.7 GB | Fits | Get file → | |
| IQ1_S | 6.2 GB | 8.8 GB | Too big | Get file → | |
| IQ1_M | 6.8 GB | 9.4 GB | Too big | Get file → | |
| IQ2_XXS | 7.8 GB | 10.4 GB | Too big | Get file → | |
| IQ2_XS | 8.6 GB | 11.3 GB | Too big | Get file → | |
| IQ2_S | 8.9 GB | 11.5 GB | Too big | Get file → | |
| IQ2_M | 9.7 GB | 12.3 GB | Too big | Get file → | |
| Q2_K_S | 10.0 GB | 12.6 GB | Too big | Get file → | |
| Q2_K | 10.7 GB | 13.4 GB | Too big | Get file → | |
| IQ3_XXS | 11.2 GB | 13.8 GB | Too big | Get file → | |
| IQ3_XS | 11.9 GB | 14.6 GB | Too big | Get file → | |
| Q3_K_S | 12.6 GB | 15.2 GB | Too big | Get file → | |
| IQ3_S | 12.6 GB | 15.3 GB | Too big | Get file → | |
| IQ3_M | 12.9 GB | 15.5 GB | Too big | Get file → | |
| Q3_K_M | 14.0 GB | 16.6 GB | Too big | Get file → | |
| Q3_K_L | 15.1 GB | 17.8 GB | Too big | Get file → | |
| IQ4_XS | 15.5 GB | 18.1 GB | Too big | Get file → | |
| Q4_0 | 16.4 GB | 19.1 GB | Too big | Get file → | |
| Q4_K_S | 16.5 GB | 19.1 GB | Too big | Get file → | |
| Q4_K_M | 17.5 GB | 20.1 GB | Too big | Get file → | |
| Q4_1 | 18.1 GB | 20.8 GB | Too big | Get file → | |
| Q5_K_S | 19.9 GB | 22.6 GB | Too big | Get file → | |
| Q5_K_M | 20.5 GB | 23.1 GB | Too big | Get file → | |
| Q6_K | 23.7 GB | 26.3 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (6.15 GB file, 8.8 GB memory at 8 k context) or IQ1_M (6.76 GB file, 9.4 GB memory). With 12 GB VRAM, IQ2_XS (8.63 GB file, 11.3 GB memory) or Q2_K (10.72 GB file, 13.4 GB memory) are suitable. Larger VRAM allows higher‑quality quants such as Q3_K_M (13.96 GB file, 16.6 GB memory).
What GPU do you need for IBM_granite 4.1 30b_Abliterated?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 2.7 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 2.7 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 2.7 GB in use).
- 12 GB VRAM: the best quant that fits is IQ2_S (8.9 GB file, about 11.5 GB in use).
- 16 GB VRAM: the best quant that fits is IQ3_M (12.9 GB file, about 15.5 GB in use).
- 24 GB VRAM: the best quant that fits is Q5_K_M (20.5 GB file, about 23.1 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (23.7 GB file, about 26.3 GB in use).
Which IBM_granite 4.1 30b_Abliterated quant should you choose?
Q4_K_M is the usual starting point for IBM_granite 4.1 30b_Abliterated: it balances file size and quality well. Here it is a 17.5 GB file.
If memory is tight, the smallest file is the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about IBM_granite 4.1 30b_Abliterated
How much VRAM does IBM_granite 4.1 30b_Abliterated need?
About 20.1 GB at Q4_K_M with an 8K context window. That is the 17.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run IBM_granite 4.1 30b_Abliterated on an 8 GB GPU?
Yes, with the smaller files: the 0.0 GB file needs about 2.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of IBM_granite 4.1 30b_Abliterated GGUF?
It depends on the quant: file is 0.0 GB, IQ1_S is 6.2 GB, IQ1_M is 6.8 GB, IQ2_XXS is 7.8 GB, IQ2_XS is 8.6 GB, IQ2_S is 8.9 GB and 18 more.
What context length does IBM_granite 4.1 30b_Abliterated support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.
Where can I download IBM_granite 4.1 30b_Abliterated GGUF?
The files are on Hugging Face in the mradermacher/IBM_granite-4.1-30b_Abliterated-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does IBM_granite 4.1 30b_Abliterated use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run this model on 8 GB VRAM?
Yes, use IQ1_S or IQ1_M quants; they require approximately 8.8–9.4 GB memory at 8 k context.
Does the model support vision tasks?
No, the model is purely textual and has no vision capabilities.
What license applies to this model?
The model is released under the apache-2.0 license.